DESCRIPTION (Applicant's abstract): This application seeks funding to continue a successful research program of HIV prevention cost-effectiveness assessment. In the initial period of funding, our research established that small-group format, sexual risk reduction interventions to prevent the spread of HIV can be highly cost-effective. However, small-group and individually focused interventions are necessarily limited in scale and scope. Community-level interventions are also needed to facilitate large-scale reductions in HIV transmission in at risk populations. At present, relatively little is known about the cost-effectiveness of community-level interventions, which attempt to alter the behavioral patterns of entire communities rather than just a small number of individuals. Even less is known regarding how local circumstances -- such as differences in the prevailing wage rate and cost of living, and differences in the composition and structure of the target community -- affect the costs and outcomes of HIV prevention programs. We propose to conduct cost-effectiveness and cost-utility analyses of three community-level sexual behavior risk reduction interventions, including two multisite trials that were implemented at locations across the U.S. (one targeted low-income, predominately African-American women living in low-income housing developments in five cities, while the other focused on the gay-bar-going communities of four cities). These multisite analyses will allow us to: 1) thoroughly examine the economic efficiency of community-level HIV prevention interventions; 2) explicate the local factors that influence the cost-effectiveness of these interventions; and 3) quantify the variability observed in the costs and outcomes associated with conducting the "same" intervention under disparate conditions. Although "efficacy" information of the sort provided by the proposed analyses -- which assess the cost-effectiveness of interventions delivered under carefully controlled research conditions -- is critical for establishing the economic efficiency of these programs under "ideal" circumstances, information is also needed regarding how cost-effective these interventions are when implemented under "real-world" conditions by local health departments, community-based organizations (CBOs), and other service providers. To examine this issue further, we will perform a cost-effectiveness analysis of a community-level sexual risk reduction intervention for men who have sex with men that was conducted by the Kentucky Department of Health in collaboration with a local CBO. Because this intervention was based on the same psychosocial model as the multisite gay bar intervention, this analysis will enable us to compare the costs and consequences of a particular intervention type when implemented by a local agency rather than by a highly trained research team. Upon completion of the analyses we will summarize and widely disseminate the results.